knowledge-base
Pass
Audited by Gen Agent Trust Hub on May 4, 2026
Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill defines an ingestion process in
scripts/kb-ingest.tsthat reads markdown files from the local filesystem and transmits their content to the WaniWani KB API for indexing. This behavior is consistent with the skill's stated purpose of setting up a remote knowledge base. - [COMMAND_EXECUTION]: The skill directs the agent to use shell commands including
bun run kb:ingestandbun run buildto execute the generated ingestion script and verify the project state. - [PROMPT_INJECTION]: The
faq.tstool introduces an indirect prompt injection surface by retrieving and displaying content from ingested markdown files without boundary markers. 1. Ingestion points: Markdown files located inlib/{MCP_NAME}/knowledge-base/knowledge/are processed byscripts/kb-ingest.ts. 2. Boundary markers: No delimiters or protective instructions are used in the search tool's output to separate search results from system instructions. 3. Capability inventory: The skill usesclient.kb.ingestfor data upload andclient.kb.searchfor data retrieval. 4. Sanitization: No validation or sanitization of the markdown content is performed before ingestion or retrieval.
Audit Metadata